SCoRE: Streamlined Corpus-based Relation Extraction using Multi-Label Contrastive Learning and Bayesian kNN

Luca Mariotti, Veronica Guidetti, Federica Mandreoli · 2025

Read SCoRE: Streamlined Corpus-based Relation Extraction using Multi-Label Contrastive Learning and Bayesian kNN by Luca Mariotti, Veronica Guidetti, Federica Mandreoli (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-06895.txt.

Description

Luca Mariotti, Veronica Guidetti, and Federica Mandreoli introduce a system called SCoRE in this 2025 paper. The system pulls facts about links between things from text, a task used to grow knowledge graphs. It is built to work with little training data and to swap in large language models with ease.

SCoRE pairs a learning method that groups like cases with a simple neighbor-based rule for labels. It needs no fine-tuning of the language model. The authors also offer two new scores for judging results and a new test set called Wiki20d. Tests on five data sets show scores that match or beat prior work while using far less energy.

Catalog record

CREATORLuca Mariotti, Veronica Guidetti, Federica Mandreoli
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE71 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2507-06895

Browse the full library · About the Tower of Babel